Translation policy based on IPv6 protocol stack

We have already talked about the dual-stack protocol and tunnel protocol in our discussion of the networking proposal for coexistence of two networks. This is also the mainstream strategy. Next, let's introduce Translation Strategies to you. There may be fewer people who know about them. Here, we will focus on the translation policies related to the IPv6 protocol stack.

Translation Strategy

During the network transition period, it is impossible for all hosts or terminals to be upgraded to support dual-stack. in the network, communication between pure IPv4 hosts and pure IPv6 hosts is required, because of the differences in the protocol stack, it is natural to translate and convert these protocols. Protocol translation involves two aspects: Translation between IPv4 and IPv6 protocol layers, and translation between IPv4 and IPv6 protocol stack applications. Translation Strategies can correspond to a variety of implementation technologies, NAT-PT and TRT are mainly used in the network convergence layer, while BIA and BIS are mainly proposed for Host terminal.

1) NAT-PT

NAT-PT gateway can realize the mutual conversion between IPv4 and IPv6 protocol stacks, including network layer protocol, transport layer protocol and some application layer protocol, the original protocols can interwork with the new ones without modification, but the technology has some restrictions on application:

◆ In terms of topology, the conversion of two-way data packets in a session must be completed on the same vro. Therefore, the address/protocol conversion method is more suitable for networks with only one vro egress;

◆ Some protocol fields cannot completely maintain the original meaning during conversion.

2) TRT

Transmission Relay converter ("TRT" Transport Relay Translator) is suitable for communication between pure IPv6 networks and pure IPv4 networks. The TRT system is located between a pure IPv6 host and a pure IPv4 host, and can be used to translate TCP, UDP), IPv6, TCP, UDP), and IPv4 data. Transmission relay can be divided into TCP relay and UDP relay.

The biggest difference between TRT and NAT-PT is that as a relay, TRT communicates with both sides in the TCP/UDP layer as a proxy. For example, TCP relay establishes a TCP connection with both sides of TCP communication, all TCP communications between the two parties are forwarded by TCP relay, while the NAT-PT only serves as a translation and does not act as a proxy communication.

3) BIS

BIS technology is used to monitor data streams between TCP/IP modules and nic drivers by adding several modules to the dual-stack host, such as translator, extended domain name parser, and address er, and translate the corresponding IPv4 and IPv6 protocol stack packets. When communicating with other IPv6 hosts, some IPv4 addresses are allocated to the corresponding IPv6 hosts on this host. These addresses are only used within this host. In addition, this allocation process is automatically completed through the DNS protocol. The host can use an existing IPv4 application to communicate with other IPv6 hosts, making it a dual-stack host that supports both IPv4 and IPv6 applications, this expands the application of double-stack hosts. In addition, the BIS mechanism can coexist with other conversion mechanisms.

4) BIA

BIA technology adds an API translator between the Socket API module of the dual-stack host and the TCP/IP module, which can translate between IPv4 Socket API functions and IPv6 Socket API functions, this mechanism simplifies the translation between IPv4 and IPv6, and does not require translation of IP addresses.

The dual-stack host using BIA assumes that both TCP/IPv4 and TCP/IPv6 protocol stacks exist on the local node. When the IPv4 application on the dual-stack host communicates with other IPv6 hosts, the API translator detects the Socket API function in the IPv4 application, and call the Socket API function of the IPv6 protocol stack to communicate with the IPv6 host, and vice versa.